Averaged Velocity Fields Research Articles

Characterizing the Molecular Gas in Infrared Bright Galaxies with CARMA

We present the CO(1–0) maps of 28 infrared-bright galaxies from the Great Observatories All-Sky Luminous Infrared Galaxy Survey (GOALS) taken with the Combined Array for Research in Millimeter Astronomy (CARMA). We detect 100 GHz continuum in 16 of the 28 CARMA GOALS galaxies, which trace both active galactic nuclei (AGNs) and compact star-forming cores. The GOALS galaxies show a variety of molecular gas morphologies, though in the majority of cases the average velocity fields show a gradient consistent with rotation. We fit the full continuum spectral energy distributions (SEDs) of each of the sources using either magphys or SED3FIT (if there are signs of an AGN) to derive the total stellar mass, dust mass, and SFRs of each object. We adopt a value determined from luminous and ultraluminous infrared galaxies (LIRGs and ULIRGs) of α CO = 1.5−0.8+1.3 M ⊙ (K km s−1 pc2)−1, which leads to more physical values for f mol and the gas-to-dust ratio. Mergers tend to have the highest gas-to-dust ratios. We assume the cospatiality of the molecular gas and star formation and plot the CARMA GOALS sample on the Schmidt–Kennicutt relation, where we find that they preferentially lie above the line set by normal star-forming galaxies. This hyper-efficiency is likely due to the increased turbulence in these systems, which decreases the freefall time compared to star-forming galaxies, leading to “enhanced” star formation efficiency. Line wings are present in a non-negligible subsample (11/28) of the CARMA GOALS sources and are likely due to outflows driven by AGNs or star formation, gas inflows, or additional decoupled gas components.

Two-phase flows through porous media described by a Cahn–Hilliard–Brinkman model with dynamic boundary conditions

We investigate a new diffuse-interface model that describes creeping two-phase flows (i.e., flows exhibiting a low Reynolds number), especially flows that permeate a porous medium. The system of equations consists of a Brinkman equation for the volume averaged velocity field and a convective Cahn–Hilliard equation with dynamic boundary conditions for the phase field, which describes the location of the two fluids within the domain. The dynamic boundary conditions are incorporated to model the interaction of the fluids with the wall of the container more precisely. In particular, they allow for a dynamic evolution of the contact angle between the interface separating the fluids and the boundary, and for a convection-induced motion of the corresponding contact line. For our model, we first prove the existence of global-in-time weak solutions in the case where regular potentials are used in the Cahn–Hilliard subsystem. In this case, we can further show the uniqueness of the weak solution under suitable additional assumptions. We further prove the existence of weak solutions in the case of singular potentials. Therefore, we regularize such singular potentials by a Moreau–Yosida approximation, such that the results for regular potentials can be applied, and eventually pass to the limit in this approximation scheme.

Integrated positron emission particle tracking (PEPT) and X-ray computed tomography (CT) imaging of flow phenomena in twisted tape swirl flow

A combined positron emission particle tracking (PEPT) and X-ray computed tomography (CT) technique is presented, and its utility is demonstrated through investigation of flow in a pipe with twisted tape swirl insert with varying flow conditions (diameter-based Reynolds numbers 16,300–63,300). A description of this technique is given, as well as data handling practices used to relate geometric information captured by CT to fluid flow data gathered via PEPT. It is found that the CT component is readily capable of capturing the stainless steel insert geometry in this present system, but the use of combined plastic and metal materials leads to artifacts in imaging of the plastic surface. Nonetheless, CT data are related to PEPT flow measurements, and average velocity fields are calculated via a pseudo-framing and interpolation scheme and used to visualize and interrogate key flow phenomena within the system. Radial velocity profiles of the mean flow characteristics are seen to collapse to a nearly common form across all flow conditions considered. Helical vortices are seen propagating through the flow field, generated by bypass flow around the gap between the insert and pipe wall, with additional coherent secondary flow structures seen in the higher Reynolds number cases. These findings enhance the understanding of the mixing mechanisms in these swirl flows and encourage the continued development of PEPT-CT methodologies for 3D flow measurements in optically inaccessible systems.

Exploration of microscopic physical processes of Z-pinch by a modified electrostatic direct implicit particle-in-cell algorithm

Abstract For investigating efficiently the stagnation kinetic-process of Z-pinch, we develop a novel modified electrostatic implicit particle-in-cell algorithm in radial one-dimension for Z-pinch simulation in which a small-angle cumulative binary collision algorithm is used. In our algorithm, the electric field in z-direction is solved by a parallel electrode-plate model, the azimuthal magnetic field is obtained by Ampere’s law, and the term for charged particle gyromotion is approximated by the cross product of the averaged velocity and magnetic field. In simulation results of 2 MA deuterium plasma shell Z-pinch, the mass-center implosion trajectory agrees generally with that obtained by one-dimensional MHD simulation, and the plasma current also closely aligns with the external current. The phase space diagrams and radial-velocity probability distributions of ions and electrons are obtained. The main kinetic characteristic of electron motion is thermal equilibrium and oscillation, which should be oscillated around the ions, while that of ion motion is implosion inwards. In the region of stagnation radius, the radial-velocity probability distribution of ions transits from the non-equilibrium to equilibrium state with the current increasing, while of electrons is basically the equilibrium state. When the initial ion density and current peak are not high enough, the ions may not reach their thermal equilibrium state through collisions even in its stagnation phase.

Kármán Street-Boundary Layer Interactions In Turbulent Flow

Understanding the intricate interplay between turbulent boundary layers and wake structures caused by obstacles in the flow is key to assessing the local shear stress fields as well as the drag. These types of flows can be relevant to natural phenomena including sediment and nutrient transport. While existing literature predominantly explores the flow downstream of cylinders mounted perpendicular to bounding surfaces or downstream of cylinders positioned outside the boundary layer region, this work experimentally studies the turbulent boundary layer passing over a cylindrical obstacle positioned near and parallel to the surface of an underlying solid wall. Planar particle image velocimetry (PIV) was performed in an open recirculating water channel for three test cases at a friction Reynolds number Reτ = 1320. Key parameters investigated include the cylinder diameter relative to the boundary layer thickness (D/δ = 0.09 and 0.01) and the gap between the cylinder and the wall (G/δ = 0.09), where the range of D/δ used is smaller than most examples found in the literature. Instantaneous and averaged velocity and vorticity fields are examined based on the PIV measurements. The results in Fig. 1 show how von Kármán vortex streets and turbulent boundary layer structures interact to form unique flow features. The larger cylinder generates a clear separation downstream of its axis as well as a wake that remains coherent over a distance δ downstream. The smaller cylinder with D/δ = 0.01 generates a much narrower and weaker wake, but still shows a clear influence on time-averaged velocity profile. Future experiments will evaluate flow conditions further downstream as well as additional values of G/δ for the same D/δ values.

The Measured Flow at the Inlet of a Francis Turbine Runner Operating in Speed No-Load Condition

Abstract For Francis turbines, speed-no-load (SNL) represents one of the most detrimental operating conditions, marked by significant pressure and strain fluctuations on the runner. Mitigating these fluctuations necessitates a comprehensive understanding and characterization of the flow phenomena responsible for their generation. This paper presents an experimental investigation of the flow at the inlet of a Francis turbine runner model operating in speed-no-load condition using high-speed stereoscopic and endoscopic particle image velocimetry (PIV). The measurements are made in a radial-azimuthal plane that covers the vaneless space and a large region in the interblade channel. This study marks the first-time measurement of critical flow phenomena at this operating point, performed in the runner. Instantaneous and average velocity fields are analyzed, along with other statistical data. The results not only confirm the stochastic nature of the flow at speed-no-load but also highlight the general structure of the flow observed in other studies. The high velocity fluctuations on the suction side are associated with a backflow extending into the vaneless space and a circulation zone occasionally generated by this backflow. Both phenomena are frequently present, but fluctuate stochastically. Additionally, two other circulation zones intermittently form on the pressure side of the blades. The presence of vortices, smaller than the circulation zones, near the blade's leading edge correlates with the backflow intensity.

A new car following model based on weighted average velocity field

In real traffic, the motion status of forward moving vehicles play an important role in influencing the following vehicle’s motion behaviour. Nowadays, autonomous and electric vehicles have been incorporated with car-following mode when the control process is more complicated and more meaningful. As a result, researchers are studying car-following control strategies for vehicles to attain stability in traffic. Based on this, we propose a new car following model considering weighted average velocity field to describe the average effect of multi-vehicle interaction on the whole road. The weighted average velocity field can be regarded as the weighted average of velocities of finite number of vehicles preceding the objective vehicle with more weightage to more preceding vehicle. The stability condition is derived through stability analysis. Subsequently, the modified Korteweg–de Vries (mKdV) equation is formulated to unveil traffic congestion in the form of the kink-antikink density wave. Theoretical findings indicate that the weighted average velocity field yields larger stability regions compared to the average velocity field. Numerical simulations were conducted, and the results obtained demonstrate that the weighted average effect of multi-vehicle interactions play a crucial role in suppressing traffic jams faster than a simple average.

Parabolic turbulence k-epsilon model with applications in fluid flows through permeable media

In this work, we study a one-equation turbulence \(k\)-epsilon model that governs fluid flows through permeable media. The model problem under consideration here is derived from the incompressible Navier-Stokes equations by the application of a time-averaging operator used in the \(k\)-epsilon modeling and a volume-averaging operator that is characteristic of modeling unsteady porous media flows. For the associated initial- and boundary-value problem, we prove the existence of suitable weak solutions (average velocity field and turbulent kinetic energy) in the space dimensions of physics interest.

Particle Image Velocimetry in a Centrifugal Pump: Influence of Walls on the Flow at Different Axial Positions

Abstract For almost a century, humans have relied on centrifugal pumps for the transport of low-viscous fluids in commercial, agricultural, and industrial activities. Details of the fluid flow in impellers often influence the overall performance of the centrifugal pump and may explain unstable and inefficient operations taking place sometimes. However, most studies in the literature were devoted to understanding the flow in the midaxial position of the impeller, only with a few focusing their analysis on regions closer to solid walls. This paper aims to study the water flow in the vicinity of the front and rear covers (shroud and hub) of a radial impeller to address the influence of these walls on the fluid dynamics. For that, experiments using particle image velocimetry (PIV) were conducted in a transparent pump at three different axial planes, and the PIV images were processed to obtain the average velocity fields and profiles, as well as turbulence levels. Our results suggest that: (i) significant angular deviations are observed when the velocity vectors on the peripheral planes are compared with those on the central plane; (ii) the velocity profiles close to the border are similar to those in the middle, but the magnitudes are lower close to the hub than to the shroud; (iii) the turbulent kinetic energy on the periphery is up to eight times greater than that measured at the center. Our results bring new insights that can help propose mathematical models and improve the design of new impellers. A database and technical drawings of the centrifugal pump are also available in this paper so that other researchers can perform numerical simulations and validate them against experimental data.

A wall-modelled large eddy simulation approach based on large scale parallel computing for post stall flow over an iced airfoil

Safety evaluation for ice accretion is one of the most important jobs for the airworthiness certification of commercial aircraft. Ice accretion would induce strong performance degradation of lifting surfaces by modifying the geometry of the leading edge and the state of boundary layers, and inducing premature flow separation. In this paper, a wall-modelled large eddy simulation (WMLES) approach combined with high performance computing is presented and evaluated for numerical simulation of post stall flow at Mach number 0.12 and Reynolds number 3.5 million over an iced airfoil with angle of attack 6 degrees for business jet, which is known as the GL305/944 airfoil with horn ice. Both RANS and an improved DDES (IDDES) based on SST model are also completed to validate the results. The results are compared with the experimental data in details that includes total forces, velocity profile, averaged velocity field, RMS of pulsating velocity, et al. It is concluded that the WMLES approach presented here can greatly improve the numerical accuracy for post stall flow with large separation; Basically, WMLES can relatively accurately predict the total forces, the pressure rooftop length and the pressure recovery, and the shear layer instability induced by the horn ice.The relative error of lift coefficient for WMLES is only 0.47%, which is far less than RANS with -26.7%.

Numerical simulation of turbulent premixed flames with the conditional source-term estimation model using Bernstein polynomial expansion

Conditional Source-term Estimation (CSE) is a turbulence-chemistry interaction model similar to CMC, except that the conditional scalars are calculated from unconditional ones using an integral equation. This problem is inherently ill-posed and should be regularised. Recently, an efficient regularisation approach based on Bernstein polynomial expansion was proposed by Mahdipour and Salehi (Combust. Flame, 2022) in an a priori analysis using DNS data. This work implements this approach in a reacting flow solver, and two laboratory-scale turbulent premixed flames are simulated in the Reynolds-Averaged Navier-Stokes (RANS) context. The turbulent intensity in the first flame is low, and the results show that, unlike the conventional CSE approach, the new approach can accurately predict the flamelet conditional averages. Furthermore, the predicted averaged velocity field and major and minor species mass fractions compare favourably with the experimental measurements. The turbulent intensity in the second flame is relatively higher, and the predicted conditional averages should deviate from an unstrained laminar flame solution. The new approach can correctly predict this trend as well as the flame height in this flame. The computational cost of the new CSE approach is also substantially reduced compared to the regular CSE approach.

EXPERIMENTAL CHARACTERISATION OF THE WAKE OF A BOTTOM-MOUNTED TWO TANDEM OF CYLINDERS PLACED IN A HIGH VELOCITY AREA

Following the development of renewable marine energies, the characterization of areas with strong marine currents has become necessary. The Normandy coasts (France) are among the sites suitable for the installation of tidal turbine parks because they have significant energy potential. Projects to install machines in these sites raise questions about various factors affecting the performance of the turbines that would be placed there.
 We are currently working on the understanding of the mechanisms leading to the generation of ambient flow turbulence on the seabed. More particularly, we are interested in the impact of the complexity of the bathymetry on the organization of the wake generated. To simplify the modeling of the bathymetry of the seabed, we use generic cylindrical obstacles of rectangular section. The obstacles are placed on the bottom of the study area of ​​the Hydrodynamic Tunnel of the LUSAC laboratory and occupy its entire width. Measurements are made using Particle Image Velocimetry (PIV-2D) for a velocity flow of 3 m/s.
 In a first time, in a previous contribution, was evaluated the impact of the ratio between the Height and the Width (H/W) of the cross section of the cylinder on the organization of the near wake. Indeed, the average velocity fields obtained for six different obstacles, highlighted the modification of the organization of the flow topology. The variation of this ratio H/W leads to the modification of the length of the average vortex formation zone, implies the presence of two or even three average recirculation zones in the near wake of the cylinder and leads to the possible presence of a recirculation zone average placed upstream of the cylinder.
 Nevertheless, in the seabed the structures are not isolated. In the contrary, on the seabed we can observe successions of structures anchored on the funds, leading to the interaction of a structure with the wake generated upstream by another one structure. The experimental study with the characterization of the flow around a tandem of “long” cylindrical obstacles of square section (side H). The distance between these two cylinders is equal to 2H.
 As expected, in the present study, it can be observed the modifications of the topology of the mean flow and the distribution of the turbulent kinetic energy. Indeed, the interaction between the mean recirculation, generated downstream each cylinder, is shown.

Supersonic Flow Separation in Front of a Forward-Facing Step

A detailed experimental study has been done to understand the Shock Wave-Boundary Layer Interaction (SWBLI) over a Forward-Facing Step (FFS) of height (h) in a Mach 2.5 flow using Particle Image Velocimetry (PIV). In addition to PIV, unsteady wall pressure measurements, high-speed schlieren and surface oil flow visualization have been studied. Flow separation using surface oil flow visualization was observed around 4.1 step heights upstream of the step face. Shock motions extracted from the schlieren image sequence shows a peak frequency of approximately 1000 Hz, which corresponds to the dominant (low) frequency at which the shock oscillates; this being two orders in magnitude smaller than the characteristics frequency of the incoming boundary layer. Also, spectra of the unsteady wall pressure data shows that the shock oscillates with a peak frequency of approximately 1000 Hz. The effect of shock location (xs) from instantaneous velocity PIV fields in the cross-stream plane due to an upstream influence parameter (line-averaged boundary layer streamwise velocity, ul) and a downstream parameter (separation bubble or recirculation region area, A) is studied. Instantaneous cross-stream PIV measurements show significant variations of the separation shock location, incoming boundary layer velocity fluctuations and recirculation region. Besides, conditionally averaged velocity field based on incoming boundary layer streamwise velocity fluctuations and separation bubble area in the cross-stream plane shows that the shock location (xs) is correlated to the size of the separation bubble and not to boundary layer velocity fluctuations.

Comparative study of the effects caused by polymers, bubbles and surfactants in a turbulent flow

In the present work, the particle image velocimetry (PIV) technique was used to measure the velocity components in the direction normal and tangent to the wall to obtain average velocity fields, wall shear stresses, friction velocity, drag reduction and average deformation fields were obtained by adding bubbles (injected by electrolysis), polymer (WSR-301 polyox) and surfactant (cationic) and their bubble-polymer, bubble-surfactant and polymer-surfactant combinations at concentrations of 164 and 272 ppm in a water flow in a channel (2cm x 10cm x 160cm) with a Reynolds number of 5200. Increased levels of drag reduction were obtained when combining the techniques, for example in the bubbles with polymers (WSR - 301 polyox) combinations, drag reduction results of 82 and 93 % were obtained for the concentrations of 164 and 272 ppm respectively, While when the combinations of bubbles with surfactants were used, the results were 37 % for 164 ppm and only 16 % for 272 ppm, and for the combination of polymer with surfactant for 164 ppm the results were 47 % and for 272 ppm the drag increased by 25 %, possibly due to an incorrect preparation of the polymer or surfactant, which leads to the conclusion that the greatest synergistic benefit is presented when combining the drag reducing techniques of bubbles and polymers.

On-site aerodynamics investigation of speed skating

An aerodynamic assessment is presented of two elite skaters, each in two different skating postures, at the ice-rink Thialf in Heerenveen, the Netherlands, via on-site Ring of Fire (RoF) measurements. This experimental approach adopts stereoscopic Particle Image Velocimetry (Stereo-PIV) to measure the flow upstream and downstream of the skaters. Both skaters transit through the RoF 20 times, 10 in each skating configurations. Athlete A skates with two hands on the back and with one arm on the back and one loose. Athlete B skates with one arm loose in a normal deep sit and in an extreme deep sit. All tests are performed at a nominal skating speed of 11 m/s.Firstly, the wake velocity fields of skater A with two hands on the back are presented throughout five different phases of the skate stroke. Significant variations in the distribution of the velocity deficit downstream of the athlete are observed, which suggest corresponding variations in the skater's aerodynamic drag. These velocity fields are also compared to literature and the similarities and differences are discussed between the flow around a static skater and that in the natural skating motion. Secondly, average streamwise velocity and vorticity fields for all 4 different postures are presented and compared. It is observed that for all cases the maximum velocity deficit in the wake is in the range of 0.45 ≤ ux* ≤ 0.55 and is located behind the lower back and upper legs. Furthermore, a characteristic vortex pair is observed downstream of the skater's hips for all four skating configurations, indicating it is independent of the athlete, the posture, and skating phase.The ensemble average aerodynamic drag is evaluated via a control volume approach along the wake behind the skater, accounting for the non-uniform flow conditions prior to the skater's passage. The uncertainty of the average drag measurements from the present RoF is about 5%. The results show that the optimization of the deep sit, e.g. the trunk and knee angle, yields a reduction by 7.5% of the skater's aerodynamic drag. Conversely, the difference in drag between two arms loose and one arm loose is not statistically significant.

Effects of High‐Frequency Flow Variability on the Pathways of the Indonesian Throughflow

AbstractPrevious studies have shown the presence of strong mesoscale eddy activities in the Indonesian Seas and their influence on the transport and water mass properties of the Indonesian Throughflow (ITF), a mean flow from the Pacific Ocean to the Indian Ocean through the Indonesian Archipelago. This study explores the effects of these eddy activities, or high‐frequency flow variability (HFFV), on residence time and pathway of the ITF by conducting Lagrangian particle tracking experiments using a velocity field from an eddy‐resolving ocean general circulation model. Particles are released at key locations in the western and eastern routes of the ITF and tracked both backward and forward in time. To assess the effects of flow variability that has a time scale longer than a day but shorter than a month, the definition of HFFV in this study, we conduct parallel experiments using daily and monthly averaged velocity fields. Particle trajectories reveal the contrasting circulation characteristics of the Sulawesi and Banda Seas. HFFV in the Sulawesi Sea (in the western route) is high, causing water to circulate longer over a broader area. The longer residence time in the Sulawesi Sea helps the upwelling of the inflowing Pacific waters, especially the intermediate water masses, to rise above 300 m at the Makassar Strait, and also has the potential to allow mixing processes to modify the water mass properties of the ITF. In contrast, HFFV is much lower in the Banda Sea and has minimal effects on the ITF.

Hydrodynamic turbulence in disks with embedded planets

The vertical shear instability (VSI) is a source of hydrodynamic turbulence that can drive vigorous vertical mixing and moderate levels of accretion in protoplanetary disks, and it could be observable in the near future. With high-resolution three-dimensional numerical hydrodynamics simulations, we modeled the behavior of the VSI in protoplanetary disks with and without embedded planets. We then measured its accretion and mixing capabilities by comparing the full Reynolds stress, which includes the contribution of nonaxisymmetric features, such as spiral arms and vortices, to the Reynolds stress due to the azimuthally averaged velocity field, which can be attributed to good approximation to the VSI. We verified that the VSI can contribute to the accretion stress and showed that, depending on disk conditions, an embedded planet can coexist with or suppress VSI turbulent stress. Specifically, the presence of spiral shocks launched by a planet or planet-generated vortices can interfere with the VSI near the planet’s vicinity, with the instability recovering at large enough distances from the planet or vortex. Our results suggest that observations of VSI signatures are unlikely in disks that contain massive, nonaxisymmetric features.

Experimental Study of the Particles Influence on the Pyramid Wake within the Turbulent Boundary Layer

Particle imaging velocimetry (PIV) was used to study the near-field variation of a pyramid rough element in clear water and a liquid–solid boundary layer (thickness: 60 mm). Particles with an average diameter of 355 µm and Stokes number of 4.3 were injected into a 1:1000 mass ratio (solid particles: water) liquid–solid two-phase solution. Experiments were conducted to collect instantaneous velocity field information in the streamwise–normal direction and streamwise–spanwise direction at a Reynolds number of 8350. Then, the average velocity field and turbulence intensity of the rough element wake under single-phase and two-phase conditions were compared, and the morphology and periodicity of the shedding structure were analyzed by using proper orthogonal decomposition (POD) combined with the power spectral density function (PSD). Particles were shown to have no significant impact on the recirculation area in the streamwise–spanwise plane but did result in a reduction of the recirculation zone in the streamwise–normal plane and a 0.2h closer location of the streamline's origin to the obstacle. Along with the weakening of the upwash structure, the particle phase diminishes the velocity gradient along the span direction and turbulence intensity. Structural shedding at the top of the pyramid and near the wall occurred simultaneously, and the same shedding period was maintained. Particularly, in the first two POD modes, the energy of the shedding structure near the wall was higher than that at the obstacle tip, with a maximum energy differential of approximately 6%. The Strouhal number of the shedding structure decreased by particles from 0.217 to 0.209. The concentration distribution and degree of dispersion in the particle-laden flow illustrate different results, with lower statistics in the wake flow field.

Experimental study on near wake shedding structure of pyramid roughness element in liquid–solid two-phase turbulent boundary layer

Particle image velocimetry (PIV) on the liquid–solid turbulent boundary layer containing solid particles with an average diameter of 355 μm is used to study the shedding structure near-field of the pyramid with varied yaw angle values (0°, 15°, 30°, and 45°). We find that the recirculation zone grows larger with increasing value by comparing the average velocity field and turbulence intensity at various angles. When closed vortex lines develop behind the pyramid at about (h, 0.8 h), the center of the recirculation zone gradually stabilizes; in the meantime, the turbulence intensity rises relative to the small α values. Power spectral density function (PSD) and proper orthogonal decomposition (POD) are used to analyze the near-field. According to the findings, various values have an impact on wake shedding structures. At the top and base of the pyramid, shearing vortices occur simultaneously on a regular basis, and their incidence is sensitive to variations in α. Similar top shedding structures are observed at various angles in the POD results. Furthermore, with increasing α, the reattachment vortex is observed at the downstream edge of the pyramid. We study the behavior and statistical properties of particles at various positions, and our findings suggest that the upper flow field's particle concentration distribution dispersion is greater than that of the wake. Some particles from the vicinity of the pyramid is attracted into the wake's recirculation zone and is carried downstream by the shedding vortex.

Thermal convection instability of two miscible viscous fluids in a rotating annular Hele–Shaw cell

We investigate thermal convection instability in a system of two horizontal miscible liquid layers confined in an annular Hele–Shaw cell rotating uniformly about its axis and subjected to a radial temperature gradient. We first determine the Hele–Shaw averaged velocity field in each fluid layer by taking into account the Coriolis force. Thereafter, the linear stability analysis leads to an eigenvalue problem solved numerically by the spectral collocation method. Depending on the buoyancy number, the ratio of the stabilizing chemical density anomaly to the destabilizing thermal density anomaly, the centrifugal force gives rise to two convection regimes: the oscillating regime corresponding to single-cell convection over the entire width of the Hele–Shaw cell and the stratified regime with separate convection in each of the two layers. In the stratified regime, it turns out that the cells rotate in the same direction and, thus, only thermal coupling is dominant in the Hele–Shaw cell geometry, regardless of the value of the viscosity ratio. We show that increasing the curvature parameter of the cell has a stabilizing effect and decreases the critical Buoyancy number corresponding to the transition from the oscillatory to the stratified regime. At a low Ekman number, the Coriolis force is strongly stabilizing and has little effect on the critical buoyancy number. Moreover, the increase in the curvature parameter and the decrease in the Ekman number together cause the transition from the oscillating regime to a nearly stratified regime. The effects of fluid layer thicknesses and the ratio of kinematic viscosities on the active or passive character of a layer are also examined.